[Troxx]

Originally Posted by DeathsSilkyMist [You must be logged in to view images. Log in or Register.]

Did some longer parses. I am using a level 52 Monk.

Oh sweet Jesus, he actually did do a follow-up and collect more data! Color me impressed, I did not think it was in his character. This is definitely progress and should be applauded.

I'm not being sarcastic!

LET'S DIVE IN BOYS!

1. 3 minute parses are similar to 10 minute parses. It is incorrect to assume that a 3 minute parse is too inaccurate for DPS.

Part a) Yes your 3 minute parses are extremely similar to 10 minute parses

Part b) Not necessarily the case.

The last time I went and played around with Bloodmaw on my warrior I was giddy with joy watching the parse-o-meter. I had just won a Frostreaver (42/43). I had done the napkin math and knew it would be good for holding aggro. i wanted to know just how much more damage I would be doing now. For the first 5 minutes I had gotten exquisitely lucky with a much higher % of hits ... much higher hit averages and a disproportionate amount of triple attacks. I was super excited at the data I was seeing. By minute 6 that dps-o-meter started to get progressively more impotent. It ended up settling on a dps value towards the bottom of Bloodmaw's hp bar that was a good 25-30% lower than where it had started.

For the record, the end results were not disappointing - it was still a very big jump in DPS from the dual wield setup I was using at the time.

Clearly this was not average or to be expected. It was an anomaly. Had I cut that parse off at the 3 minute park instead of persisting through until the end I would have walked away with some very bad assumptions. I didn't cut the parse off though. I learned hard lessons over a decade ago with my defensive parsing. I had run all the way to GD just to see a comparison so I figured why not just kill the damn thing - all 20,000hp of it.

I do not want to squash the euphoria you're feeling right now and discourage you from continuing to stay more engaged with actual data gathering. Getting yourself back out there should never be discouraged. Data is great. More is always better. More will not always change the results, but sometimes it does. I fear for your ego if you truly do feel that the data sets you gathered were so similar that you think you have now proven that you never really need a parse longer than 3 minutes to know the truth.

This is why I encourage you to download gamparse. It really is a lot of fun. Collect your data in the background and let that little yellow-text HUD give you consistent dopamine hits from instant feedback. The more you watch what is happening in real time the more you will realize (especially if you 2h much) that the random number generator is a fickle mistress.

Last night I tanked the Scout Charisa giant on my warrior. ZERO procs the whole fight. In one previous fight I averaged 10 ppm (procs per minute). In most fights I get the average of 3ppm (2 main 1 offhand with 255 dex). Last night I got zero.

2. Corudoth is a perfectly valid target for DPS comparisons between weapons. The difference between compared weapons is the same across all mobs. You don't need to wander across Norrath parsing random mobs to compare weapons. You only parse different mobs if you want to see how much their mitigation values offset your DPS. The mitigation offset will affect the compared weapons equally.

[You must be logged in to view images. Log in or Register.]

WHOA NELLY! Hold your horses there muchacho ....

Rather than just say "no you are wrong" (because you actually are actually .... very ... EXQUISITELY ... EXTREMELY wrong here), I'd rather take you by the hand and gently lead you to the correct answer. Heaven help you if you cannot follow the logic train here - failure to understand this concept would be imply you are actually beyond hope. I legitimately believe you are smart enough - let us dive in!

Snaggles posted:

From 7/31/2023 on Bloodmaw per the above:

Earthcaller/Swiftwind = 74,012 over 688 seconds = 107dps average
Vyemm whip/swiftwind = 71,059 over 597 seconds = 119 dps
Vyemm whip/claw of Lightning = 71,542 over 576 seconds = 124 dps
Meljeldin = 74,358 over 605 seconds = 122 dps
Earthcaller & Tash stick = 50,571 over 612 seconds = 82dps (686 slash landed / 511 crush landed = .74489 ratio)

Bloodmaw is a green, level 20 mob with 50,000hp:

https://wiki.project1999.com/Bloodmaw

Derakor the Vindicator is a level 70 red con raid mob:

https://wiki.project1999.com/Derakor_the_Vindicator

Rather than just say

"well snaggles can do 122dps with his https://wiki.project1999.com/Meljeldin,_Bane_of_Giants to a green level 20 mob with 50,000 hp but only 73dps with the same weapon on Vindi ... well that means that whatever other weapon I parsed with on the level 20 50,000hp mob will also be scaled similarly so I can predict the preperformance of ANY weapons setup on Vindi (proportionally) by simply testing that same weapon combination on Bloodmaw. How easy!"

Start with the questions.
-The question of WHY can Snaggles manage to put out 122dps on the green mob
-WHY is that number so much higher on the level 20 green mob and so much lower than that on the level 70 raid mob?

Well, part of that is accuracy. Snaggles is significantly higher level than one of them and significantly lower than the other. From an accuracy standpoint I will agree that the degree to which you are accurate should be the same proportionally across weapon setups unless there is a difference in weapon skill caps for any of the weapons or setups in question. For Snaggles, this is not the case so we can assume that he would land/miss as many hits with his 2hs as he would with his dual wield.


The other part? That is the armor class of the mob.

-The higher the armor class the more hits are deflected to the average.
-The lower your damage intervals are deflected, the greater the proportion of the damage that you do will be from the damage bonus.
-Vindi has SIGNIFICANTLY HIGHER ac than Bloodmaw.
-The greater proportion of baseline damage you ARE doing to the mob from damage bonus, the higher the relative importance of damage bonus.

If that is the case, let us go back to snaggles parses:

From 7/31/2023 on Bloodmaw per the above:

Earthcaller/Swiftwind = 74,012 over 688 seconds = 107dps average
Vyemm whip/swiftwind = 71,059 over 597 seconds = 119 dps
Vyemm whip/claw of Lightning = 71,542 over 576 seconds = 124 dps
Meljeldin = 74,358 over 605 seconds = 122 dps
Earthcaller & Tash stick = 50,571 over 612 seconds = 82dps (686 slash landed / 511 crush landed = .74489 ratio)

On Bloodmaw, Vyemm whip and Claw of lightning won by a hair. Let us again go back and look at a now fairly old post in this thread:

Originally Posted by Troxx [You must be logged in to view images. Log in or Register.]

At 60 a 2handed weapon with a delay of 30 (tstaff) for example will have a chance to apply 29 damage bonus per hit at an interval of 3 seconds (1.5 seconds at max haste).

You would need a 1hander with a starting delay of 11 (before haste) to have greater dmg bonus potential. A delay of 12 isn’t fast enough.

No such weapon exists.

Admittedly the delay of 30 is quite literally THE sweet spot. Let’s use another example.

For a 2handed weapon with a delay of 40 (IFS) … it’s 34 dmg per 4 seconds.

You would need a 1 hander (main hand) with a delay of 13 to have mathematical parity with the dmg bonus potential of the IFS and a delay of 12 to beat it.

Lets look at the dmg bonus per delay/level tables together shall we:

https://lucy.allakhazam.com/dmgbonus.html?setcookie=1

https://wiki.project1999.com/Meljeldin,_Bane_of_Giants

This weapon has a delay of 38. At level 60 a delay of 38 gives you a damage bonus of 32. This damage bonus *chance* of 32 is applied every 3.8 seconds.

https://wiki.project1999.com/Silver_Whip_of_Rage

This weapon has a delay of 19. At level 60 a delay of 19 gives you a flat damage bonus of 11 (all 1handers do). This damage bonus *chance* of 11 is applied at precisely half the speed of the 2hander ... every 1.9 seconds.

For every 3.8 seconds (unhasted) the 2hander applies a damage bonus chance potential of 32. In the same exact time, the 1hander applies a damage bonus potential of 22.

That is nearly 1/3 less.

Now for a level 20 mob (trivial compared to the player level) with AC values more similar to that which is seen on group xp trash from kunark, the combinations of the 13/19 Vyemm whip RATIO and the 14/18 RATIO seem to be such that they overcame this dmg bonus advantage that the 45/38 2hander had.

But mob AC on Vindi is substantial enough to drop an eye-popping 122 dps down to something more like 73 which you might otherwise expect from the 2hander.

Can you not see how and why the 2hander is going to pull ahead?

And bear in mind that I am making the MASSIVE assumption that Snaggles has 41% worn haste other than https://wiki.project1999.com/Claw_of_Lightning. If he does not have another 41% haste item and is, say, falling back on a 34% haste belt or 36% haste cloak and has the PoSky haste cloak for those parses .... 84 or 86% combined haste vs 91% combined haste alone is enough to close the gap seen on Bloodmaw completely. If he had solid ench haste, then it would be irrelevant.

But this is why we have to control for or at least clearly define all variables.

3. Using math is perfectly valid and accurate. Using the term "napkin math" to claim math isn't useable is nonsense.

Yes and no. When we say the word "napkin math" it is because we are straying from known values and hard collected real world data. Napkin math is, by nature, extrapolative and starts to become theoretical. The basic value of 'napkin math' comes from the fact that it is grounded in observational data ... but you have to be cautious assuming that just because something was observed once (or twice or however many times) that it will mathematically carry over to all situations.

Heck, the above situation is the PERFECT example.

https://wiki.project1999.com/Corudoth

That's a LEVEL 5 turtle with 200k hp. You have collected data sets that imply that offensive parses on this level 5 turtle are very shockingly similar to an XP group (relatively high level) xp mob in Frontier Mountains. Those fights happened. You have the numbers. You even video taped it (why you insist on doing this is beyond me - I do not recall anyone ever asking for video evidence). I will agree that the parses are similar enough that I am comfortable using "napkin math" to then infer that dps done vs Corudoth with a particular setup will be similarly ... VERY SIMILAR to what you might expect off that giant in FM.

Where napkin math runs into problems is if you blindly assume that the same proportionality exists between Corudoth and ... Vindi, Avatar of War, Kael arena mobs (notoriously high ac for their level) etc etc.

There is value in napkin math. But when napkin math is used improperly you're going to get stupid irrelevant results. This was the problem with you using napkin math of you soloing a damn level 40 frog in seb. You solo'd a level 40 frog in seb using dots. You did actually do that amount of dps to that mob between you and your pet. That was fact. You had the logs. You had the videos. YOU DID A THING. You and your pet did enough damage in a unit of time that you knew you had done XYZ dps. You then said "hey shamans can be expected to do XYZ dps all the time and in all circumstances" ... and we were like ... 'uh no, that's not how it works'. The conversation was regarding who was going to be the best 4th in a group that already had a cleric and 2 charming enchanters. Mobs would be dying quickly. Those spells aren't exactly quick to cast and you were not actually going to average the same thing in a fast 20 second or less fight that you had the luxury of doing when it was just you (and only you and your pet) with the full hp bar of the mob to play with. Necromancers feel this pain as well. Dots in this era are just not very cool for xp groups where mobs die pretty quickly.

"Napkin math" is hugely problematic when it is divorced from reality and the person using it is to hyper-focused on the napkin math results to step back and think clearly.

This has been a consistent problem for you DSM. I am not making fun of, mocking, or trying to disparage you right now DSM. For once I actually am not. I am trying to give you honest feedback. This is a problem for you. Whether you have the clarity of mind to see/acknowledge this and have some personal growth ... whether you just get angsty/pissy and double down ... well that is up to you.

"Napkin math" is math. IT IS MATH. But it is math that takes a known set of variables and from there makes attempts to extrapolate and theorize - to predict outcomes in other situations. It is a wonderful thing. This phenomenon (whether we are using math at all) is one of the things we as humans can do that separates us from lower organisms. Other members of the animal kingdom do it too to varying degrees but we humans do it best.

So napkin math. Maybe this works out. Maybe it does not. In the case of weapon performance in project 99 on a level 20 or a level 5 green mob with low ac vs weapon performance on a level 70+ with very high ac? No. You have to understand actual game mechanics and understand the breakpoints.

Does this make sense or am is it going over your head? I'm not trying to be snarky or cynical ... this is just a really long post and if I've lost you I am willing to patiently circle back around.

I am confident Bcbrown is following every word I'm saying and giggling at how un-erudite my verbiage is. He is mathematically a whole heck of a lot more literate than I am.

Note that this parse ran longer than the 1h parse. I stopped parsing the 2h starting at this message, so the parses between 1h and 2h are of equal length. Feel free to use the extra values for checking damage roll distributions if you want.

I actually think the parses you obtained were obtained well. I applaud you for going back out there and getting them. You are welcome to go collect another 10, 20, 60 minutes more of parses on the same targets - but I don't think you're going to get wildly different results for your level, these weapons, and these particular mobs.

As you can see, Epic Fist + SoS does better than IFS without fistweaving at level 52. IFS pulls ahead a bit with fistweaving, assuming you do it all the time and correctly.

Against these 2 mobs and at level 52 yes very much so. Knowing that your damage bonus levels (9 for the 1.6 delay and 18 for the 4.0 delay) for your specific level still favor the epic fist's primary hand delay of 16 - this will mean that on high AC mobs ... AT LEVEL 52 ... they could be expected to perform even better than they did on the 2 mobs you chose to parse this time.

Point is if you were to take your 52 monk and do vindi a few times at level 52, you would probably see that gap widen enough that fist + SoS would way outperform IFS + fistweaving.

At level 52.

That will change as you level up.

52: Epic fist dmg bonus 9 - IFS dmg bonus 18
53: Epic fist dmg bonus 9 - IFS dmg bonus 19
54: Epic fist dmg bonus 9 - IFS dmg bonus 20
55: Epic fist dmg bonus 10 - IFS dmg bonus 24
56: Epic fist dmg bonus 10 - IFS dmg bonus 25
57: Epic fist dmg bonus 10 - IFS dmg bonus 27
58: Epic fist dmg bonus 11 - IFS dmg bonus 29
59: Epic fist dmg bonus 11 - IFS dmg bonus 31
60: Epic fist dmg bonus 11 - IFS dmg bonus 34

-Between now and 60 your epic fist dmg bonus will increase by about 23%.
-Between now and 60 your IFS dmg bonus will increase by 89%
-At level 60 you will receive mainhand triple attack. 100% of your non-fistweaving strikes are now subject to a 3rd potential strike. When dual wielding a big chunk of your damage will receive no bonus at all.

This is why you should parse 1h and 2h weapons before assuming 2h is better, especially at lower levels when the damage bonus on 2h is smaller.

This is why everyone should actually parse everything and frequently before making assumptions. I am glad you are finally starting to realize this. It is what we have been telling you all along.

You showed (and fairly thoroughly) that a level 52 monk with your gear on these mobs will do more lazy autoattack dps with your specific dual wield setup compared to the IFS specifically. I do not find your results shocking at all. I have run the parses at 60. I know the game and have been aware that you are still at the level that a very fast weapon such as epic fist holds an advantage over the very slow IFS - it's damage bonus has not yet matured. It has the same damage bonus potential at level 52 that you would see on a 1 handed weapon with a mainhand delay of 20. Epic fist has a delay of 16. Had you compared it to the Tstaff? Tstaff would have won. I am pretty confident in that. If you have a buddy who will lend you a Tstaff at level 52 to give it a whirl that would be great and I would happily buy you a beer if my prediction is wrong.

You also showed that if you are not lazy the IFS still wins by using a known trick to sneak in an offhand punch. Do most monks do it? I did at some times, I did not at other times. I was more inclined to do it when I was soloing vs chain pulling in a group. APM mental fatigue is a thing for most of us.

Thank you for getting parses. Now we can circle back around to:

Originally Posted by Troxx [You must be logged in to view images. Log in or Register.]

-the fact that you will actually take less damage when soloing/tanking from having a fraction of the ripostes from the mob (every little bit adds up)
-damage shields do exist

If a young monk wants all the data they certainly are interested in the above. To whatever degree a monk cares about riposte or dmg shield damage, you're going to be taking (and I am estimating here) at least 300-400% more of it by using your dual wield.

For the monk that is only hitting from behind and being a dps class. If you want to are happy to play lazy and leave easy dps on the board, just use your dual wield. If you are soloing or tanking, IFS is the clear winner. You'll take a lot less damage over time while at the same time having the potential to do more dps than your dual wield setup.

Cool beans.

Now level to 60 and watch the world you now know and have parsed flip on it's head.

That's why I said:

Originally Posted by Troxx [You must be logged in to view images. Log in or Register.]

This is why dual wielding for monks is so generally lame at this point. There has to be a massive (BiS endgame dual wield setups) ratio advantage to compete (beat) with 38/40 or 29/30 EC monk 2hander. At that point they only win via massive ratio superiority. If you have access to those weapons, you also have access to ToV or VP 2handers (and probably many for cheaper dkp) with ratios that far outclass the EC gear.

The higher the ac, the more relative importance the portion of your damage from damage bonus.

Math is math. It really isn’t complicated.

I will admit this post was made strictly from the perspective of level 60. But lets be honest ... the game has been out for 25+ years and this project is so old that I personally have leveled every class in the game other than rog/sk/wiz/ench to 60 since finding the server 10 years ago. This includes a lot of really casual play and at least 4-5 non-consecutive years not playing at all.

Nearly everyone is working on their N'th alt ... many of which have endgame velious twink gear.